Radiation therapy is rarely used in the treatment of osteosarcoma. The dose required to provide effective local control (80-100 Gy) causes too much damage to surrounding normal tissue to be delivered routinely and safely. Despite the introduction of effective systemic chemotherapy, osteosarcoma patients who cannot achieve a complete surgical resection are never cured. Thus, there is an urgent need for the development of effective local control measures for patients whose tumors are unresectable. Our group has been studying a radiopharmaceutical, 153Sm-EDTMP, for this purpose. This agent can deliver cytotoxic radiation to tumors with exquisite precision, sparing surrounding normal tissue more than 2 mm away. So far, efficacy has been limited because of difficulties delivering sufficient radiation by this means. In the proposed clinical trial, we are testing the hypothesis that tandem treatments with 153Sm-EDTMP can deliver sufficient radiation to allow more tolerable doses of external beam radiotherapy to be administered, and that if the biological equivalent dose of the combination treatment reaches 80-100 Gy, effective local control can be accomplished. Our project will also test the hypothesis that calculating the tumor absorbed dose after administration of a small activity of 153Sm-EDTMP will allow us to calculate a larger activity to administer that will allow us to precisely reach our target dose of 80-100 Gy. Our published work has demonstrated a linear relationship between administered activity and absorbed dose in each patient (though not between patients) after administration of fixed amounts of activity. Our current treatment plan will allow variation in the second administration, with the administered activity to be based on dosimetry calculations after the first treatment. Finally, in previous clincal trials we found little correlation between tumor absorbed dose and clinical response. It is not clear what characteristics account for this variable response to treatment. Hypoxic tumors are less responsive to standard radiation than normoxic tumors, probably because part of the cytotoxicity of radiation therapy involves the generation of oxygen free radicals. We will use a novel technique, [18F]-MISO PET to test the hypothesis that hypoxia makes tumors resistant to radiopharmaceuticals, just as it contributes to resistance to external beam radiotherapy. In summary, this work will build on our prior experience treating osteosarcoma patients with 153Sm-EDTMP, using individualized dosing, combining the radiopharmaceutical with standard external beam radiotherapy, and evaluating the ability of [18F]-MISO PET to predict response to therapy. These results will allow us to continue to develop this promising agent for a group of patients not normally treated with radiation. In addition, this treatment approach can provide the basis for combining radiopharmaceuticals with external beam radiotherapy for other tumor types as well.